Curating narrative experiences through automated content compilation

Abstract

A content compilation system includes a computing platform having a hardware processor and a memory storing a software code configured to provide an editorial interface. The hardware processor executes the software code to receive compilation authoring data via the editorial interface, identify one or more end-user(s) for receiving a content compilation, access a consumption profile of the end-user(s), obtain, using the consumption profile and a first authoring criterion in the compilation authoring data, content items from one or more content sources. The software code further aggregates, using a second authoring criterion in the compilation authoring data, the content items into content subsets, groups, using a third authoring criterion, at least some of the content subsets to produce the content compilation, computes a desirability score predicting the desirability of the content compilation to the end-user(s), and provides, when the desirability score satisfies a predetermined threshold, the content compilation to the end-user(s).

What is claimed is: 
     
       1. A content compilation system comprising:
 a computing platform having a hardware processor and a system memory; 
 a compilation authoring template generated using a trained-machine learning model and stored in the system memory, wherein the trained machine learning model is configured to generate the compilation authoring template implementing an editorial style of a human editor; and 
 a software code stored in the system memory; 
 the hardware processor configured to execute the software code to:
 obtain a plurality of content items from at least one content source; 
 aggregate the plurality of content items, based on relevance to one another, into a plurality of content subsets, according to a respective one of a plurality of narrative stories of each of the plurality of content subsets; 
 produce a content compilation, using the compilation authoring template and by concatenating the plurality of content subsets, to create a narrative experience having a narrative arc determined by a sequence in which the plurality of content subsets are combined to produce the content compilation; and 
 provide the content compilation to one or more end-users. 
 
 
     
     
       2. The content compilation system of  claim 1 , wherein the content compilation is individualized for the one or more end-users based on a profile of the one or more end-users. 
     
     
       3. The content compilation system of  claim 1 , wherein the content compilation is unique to each of the one or more end-users. 
     
     
       4. The content compilation system of  claim 1 , wherein the hardware processor is further configured to execute the software code to:
 determine a first desirability score predicting a first desirability of the content compilation to the one or more end-users. 
 
     
     
       5. The content compilation system of  claim 4 , wherein the first desirability score comprises an end-user relevance score. 
     
     
       6. The content compilation system of  claim 4 , wherein the hardware processor is further configured to execute the software code to:
 determine a second desirability score predicting a second desirability of a second content compilation to the one or more end-users; and 
 provide the second content compilation to the one or more end-users, when the second desirability score satisfies a predetermined threshold. 
 
     
     
       7. The content compilation system of  claim 1 , wherein the hardware processor is further configured to execute the software code to:
 receive feedback data, wherein the feedback data rates an actual desirability of the content compilation to the one or more end-users; and 
 modify, using the trained machine learning model and the feedback data, the compilation authoring template to improve a performance by the content compilation system. 
 
     
     
       8. A method for use by a content compilation system including a computing platform having a hardware processor and a system memory storing a software code and a compilation authoring template generated using a trained machine learning model, wherein the trained machine learning model is configured to generate the compilation authoring template implementing an editorial style of a human editor, the method comprising:
 obtaining, by the software code executed by the hardware processor, a plurality of content items from at least one content source; 
 aggregating, by the software code executed by the hardware processor, the plurality of content items, based on relevance to one another, into a plurality of content subsets, according to a respective one of a plurality of narrative stories of each of the plurality of content subsets; 
 producing a content compilation, by the software code executed by the hardware processor, using the compilation authoring template and by concatenating the plurality of content subsets, to create a narrative experience having a narrative arc determined by a sequence in which the plurality of content subsets are combined to produce the content compilation; and 
 providing, by the software code executed by the hardware processor, the content compilation to one or more end-users. 
 
     
     
       9. The method of  claim 8 , wherein the content compilation is individualized for the one or more end-users based on a profile of the one or more end-users. 
     
     
       10. The method of  claim 8 , wherein the content compilation is unique to each of the one or more end-users. 
     
     
       11. The method of  claim 8 , further comprising:
 determining, by the software code executed by the hardware processor, a first desirability score predicting a first desirability of the content compilation to the one or more end-users. 
 
     
     
       12. The method of  claim 11 , wherein the first desirability score comprises an end-user relevance score. 
     
     
       13. The method of  claim 11 , further comprising:
 determining, by the software code executed by the hardware processor, a second desirability score predicting a second desirability of a second content compilation to the one or more end-users; and 
 providing the second content compilation to the one or more end-users, by the software code executed by the hardware processor, when the second desirability score satisfies a predetermined threshold. 
 
     
     
       14. The method of  claim 8 , further comprising:
 receiving, by the software code executed by the hardware processor, feedback data, wherein the feedback data rates an actual desirability of the content compilation to the one or more end-users; and 
 modifying, by the software code executed by the hardware processor, using the trained machine learning model and the feedback data, the compilation authoring template to improve a performance by the content compilation system.